Coronary artery plaque volume and obesity in patients with diabetes: the factor-64 study

Abstract

Purpose: To determine the relationship between coronary plaque detected with coronary computed tomographic (CT) angiography and clinical parameters and cardiovascular risk factors in asymptomatic patients with diabetes.

Materials and methods: All patients signed institutional review board-approved informed consent forms before enrollment. Two hundred twenty-four asymptomatic diabetic patients (121 men; mean patient age, 61.8 years; mean duration of diabetes, 10.4 years) underwent coronary CT angiography. Total coronary artery wall volume in all three vessels was measured by using semiautomated software. The coronary plaque volume index (PVI) was determined by dividing the wall volume by the coronary length. The relationship between the PVI and cardiovascular risk factors was determined with multivariable analysis.

Results: The mean PVI (±standard deviation) was 11.2 mm(2) ± 2.7. The mean coronary artery calcium (CAC) score (determined with the Agatston method) was 382; 67% of total plaque was noncalcified. The PVI was related to age (standardized β = 0.32, P < .001), male sex (standardized β = 0.36, P < .001), body mass index (BMI) (standardized β = 0.26, P < .001), and duration of diabetes (standardized β = 0.14, P = .03). A greater percentage of soft plaque was present in younger individuals with a shorter disease duration (P = .02). The soft plaque percentage was directly related to BMI (P = .002). Patients with discrepancies between CAC score and PVI rank quartiles had a higher percentage of soft and fibrous plaque (18.7% ± 3.3 vs 17.4% ± 3.5 [P = .008] and 52.2% ± 7.2 vs 47.2% ± 8.8 [P < .0001], respectively).

Conclusion: In asymptomatic diabetic patients, BMI was the primary modifiable risk factor that was associated with total and soft coronary plaque as assessed with coronary CT angiography.

Trial registration: ClinicalTrials.gov NCT00488033.

Figures

Figure 1:

Coronary artery segment anatomy used in current study and adapted from that used by Miller et al (25). A, Segmental nomenclature for coronary artery territories. Not pictured are C5 (third obtuse marginal), C7 (left posterior descending), and C8 (ramus intermedius) segments. B, Full segmentation of arteries more than 2 mm in diameter. C, Segmentation in same patient shows proximal segments C1, L1, L2, and R1.

Figure 2:

Segmentation of left anterior descending artery with vascular model and coronary cross sections. Plaque subtypes are shown as follows: red = soft plaque, blue = fibrous plaque, yellow = calcified plaque. Green = luminal area. Note exclusion of ostia during segmentation. Plaque volume is the sum of all plaque component volumes.

Figure 3:

Automated plaque detection. Red = soft plaque, blue = fibrous plaque, yellow = calcified plaque, green = luminal area. A, Image in patient with minimal plaque. PVI = 6.4 mm2, soft PVI = 1.5 mm2, fibrous PVI = 3.3 mm2, calcified PVI = 1.6 mm2. B, Image in patient with large amounts of calcified plaque. PVI = 20.1 mm2, soft PVI = 2.5 mm2, fibrous PVI = 6.1 mm2, calcified PVI = 11.5 mm2. C, Image in patient with both calcified plaque (upper left) and mixed plaque (lower right). PVI = 16.9 mm2, soft PVI = 3.5 mm2, fibrous PVI = 7.1 mm2, calcified PVI = 6.2 mm2.

Figure 4:

Bar chart shows PVI and composition according to CAC score.

Figure 5:

Bar chart show percentage agreement between quartile rank of CAC score and PVI. Patients with intermediate CAC scores (second and third quartiles) show lowest agreement with PVI quartile (28% and 34%, respectively). There was a discrepancy in rank quartile classification between PVI and CAC score in 128 of 224 patients (57%) (P < .001).